TEST #921 RESULTS
05/19/2025 - 4:22 PM
ACCELERATOR
MODEL
48.7
tokens/s
generation
1.02
sec
time to first token
1161
tokens/s
prompt
381
LocalScore
HOW YOU STACK UP
Explore All ResultsQwen2.5 14B Instruct - Q4_K - Medium
SYSTEM
CPU
Intel Xeon Gold 6152 CPU @ 2.10GHz (skylake-avx512)
RAM
503.5GB
OS
Linux
Kernel Release
5.15.0-101-generic
Architecture
x86_64
Version
Cosmopolitan 3.9.7 MODE=x86_64; #111-Ubuntu SMP Tue Mar 5 20:16:58 UTC 2024
RUNTIME
Name
llamafile
Version
0.9.2
Commit Hash
a30b324
DETAILED RESULTS
TEST NAME
PROMPT
GENERATION
TTFT
pp1024+tg16
1468
tokens/s
50.8
tokens/s
717
ms
pp4096+tg256
1271
tokens/s
40.3
tokens/s
3.25
sec
pp2048+tg256
1403
tokens/s
47.1
tokens/s
1.48
sec
pp2048+tg768
1404
tokens/s
46.0
tokens/s
1.48
sec
pp1024+tg1024
1477
tokens/s
49.7
tokens/s
713
ms
pp1280+tg3072
1343
tokens/s
44.8
tokens/s
973
ms
pp384+tg1152
1313
tokens/s
52.6
tokens/s
311
ms
pp64+tg1024
365
tokens/s
54.0
tokens/s
193
ms
pp16+tg1536
409
tokens/s
53.1
tokens/s
57
ms